Kinetic analysis and numerical tests of an adaptive car-following model for real-time traffic in ITS

Abstract

Abundant real-time vehicle trajectory information provides an important guarantee for the driving safety and drivers’ decision-making in the intelligent transportation system (ITS), which improves the transport efficiency of traffic flow at the micro level. In this paper, we propose an adaptive car-following model to study the influence of the preceding vehicles’ information on the motion state of the target and the kinetic characteristics of traffic flow in the ITS, where the space headway, the velocity difference and the acceleration of preceding vehicles are considered simultaneously. According to the generalized diagram structure of road traffic, we mainly consider the two vehicles ahead and introduce adaptive parameters to control the proportion of different influencing factors. The rationality of the extended model is verified via real-world vehicle trajectory fitting experiments with the collected empirical data. Theoretical derivations are then carried out to describe the evolution of traffic flow under different scenarios, where the linear stability conditions and nonlinear analysis on small disturbance are included. Numerical tests are made to simulate the driving conditions and verify the accuracy of theoretical results. Computing the engine power of vehicles, we summarize the impact of traffic congestions and stop-and-go behaviors on the energy consumption. Diverse micro motion states and traffic flow dynamics can be revealed based on our car-following model, which will serve as both theoretical basis and experimental examples for the real-world traffic.